Reciprocal and Coordinate Regulation of Serum Amyloid A Versus Apolipoprotein A-I and Paraoxonase-1 by Inflammation in Murine Hepatocytes

【摘要】  Objectives- During inflammation, the serum amyloid A (SAA) content of HDL increases, whereas apolipoprotein A-I (apoA-I) and paraoxonase-1 (PON-1) decrease. It remains unclear whether SAA physically displaces apoA-I or if these changes derive from coordinated but inverse transcriptional regulation of the HDL apolipoprotein genes. Because cytokines stimulate the hepatic expression of inflammatory markers, we investigated their role in regulating SAA, apoA-I, and PON-1 expression.

Methods and Results- A cytokine mixture (tumor necrosis factor -, interleukin -1ß, and IL-6) simultaneously induced SAA and repressed apoA-I and PON-1 expression levels. These effects were partially inhibited in cells pretreated with either nuclear factor B (NF- B) inhibitors (pyrrolidine dithiocarbamate, SN50, and overexpression of super-repressor inhibitor B) or after exposure to the peroxisome proliferator-activated receptor- (PPAR ) ligands (WY-14643 and fenofibrate). Consistent with these findings, the basal level of SAA was increased, whereas apoA-I and PON-1 decreased in primary hepatocytes from PPAR -deficient mice as compared with wild-type mice. Moreover, neither WY-14643 nor fenofibrate had any effect on SAA, apoA-I, or PON-1 expression in the absence of PPAR.

Conclusion- These results suggest that cytokines increase the expression of SAA through NF- B transactivation, while simultaneously decreasing the expression of apoA-I and PON-1 by inhibiting PPAR activation. Inflammation may convert HDL de novo into a more proatherogenic form by coordinate but inverse transcriptional regulation in the liver, rather than by physical displacement of apoA-I by SAA.

Stimulation of cultured hepatocytes with proinflammatory cytokines increased the expression of SAA and concurrently inhibited apoA-I and PON-1 expression. Thus, changes in the regulation of these HDL apolipoproteins could account for changes in HDL composition during inflammation rather than displacement of apoA-I and PON-1 by SAA.

【关键词】  highdensity lipoprotein serum amyloid A NF B PPAR apolipoprotein A

Introduction

Chronic low grade elevation of inflammatory markers such as C-reactive proteins (CRP) and serum amyloid A (SAA) are associated with increased cardiovascular risk. 1,2 SAA are apolipoproteins that are transported predominantly in high density lipoprotein (HDL). 3,4 SAA also have a carboxy-terminal proteoglycan binding domain, 5 which may facilitate lipoprotein retention by proteoglycans in the vascular intima, thus potentially being atherogenic. 6 SAA is synthesized primarily in the liver, where expression increases in response to cytokines such as interleukin (IL)-6, IL-1ß, and tumor necrosis factor (TNF)-. 7 See page 1687

HDL plays a critical role in reverse cholesterol transport, which is believed to account for much of the atheroprotective effect of HDL. 8,9 Cholesterol removal from cells via the ABCA1 pathway is mediated primarily by apolipoprotein A-I (apoA-I). 10,11 HDL levels decrease during acute inflammation 12-15 and are low in chronic conditions associated with increased levels of inflammatory markers. 16 ApoA-I expression is inhibited by inflammatory stimuli. 17,18

Paraoxonase-1 (PON-1) is another HDL apolipoprotein. 19 PON-1 is considered to be atheroprotective and decreases after inflammatory stimuli. 14,20,21 HDL from mice exposed to cytokines has impaired capacity to inhibit low-density lipoprotein (LDL) oxidation and protect against monocyte adherence. 22 Little is known about the hepatic regulation of PON-1 expression in relation to apoA-1 and SAA.

Isolated SAA has been shown to displace apoA-I and PON-1 from HDL in vitro. 4,23,24 However, because apoA-I expression can be inhibited by inflammatory stimuli, 17,18 it is possible that alterations in HDL apolipoprotein composition during inflammation are the result of changes in the hepatic regulation of these apolipoproteins, rather than caused by physical displacement of apoA-I or PON-1 by SAA. Moreover, it is unknown whether the expression of apoA-I, PON-1, and SAA are regulated independently or coordinately.

The current study was undertaken to evaluate whether proinflammatory cytokines coordinately regulate the expression of SAA, apoA-I, and PON-1 in hepatocytes. We demonstrate that cytokines coordinately increase SAA and decrease apoA-I and PON-1. These reciprocal changes are promoted by nuclear factor B (NF- B) and repressed by the nuclear receptor peroxisome proliferator-activated receptor- (PPAR ) in a manner dependent on both these key transcriptional mediators. These observations are intriguing given that neither apoA-I nor PON-1 promoters contain NF- B sites, 25-28 and the SAA promoters do not contain PPAR response elements (PPRE), 7 further supporting the coordinated regulation of these HDL apolipoproteins.

Materials and Methods

Cell Culture and Transfection

The murine hepatoma cell line Hepa 1-6 was cultured in DMEM supplemented with 10% FBS. The murine AML12 and NMH hepatocyte cell lines were cultured in DMEM-Ham F-12 medium supplemented with 10% FBS, ITS solution (BD Biosciences), and 0.1 nmol/L dexamethasone. 29

Hepa 1-6 cells were transiently transfected with a CMV- N-I B expression vector that overexpresses a nonphosphorylatable N-terminal deletion mutant of inhibitor B (I B) and LacZ plasmids using Fugene 6 (Roche). 30 After 24 hours, cells were exposed to a mixture of cytokines (IL-6, IL-1ß, and TNF-, all at 10 ng/mL; R&D Systems) ±other reagent(s). Cells were harvested after the indicated times, and the results were normalized to LacZ expression to control for transfection efficiency. 31

Mouse hepatocytes were isolated from 8- to 10-week-old 129S1/SvImJ PPAR wild-type and PPAR -deficient mice as described. 32

Animals

To study the effect of inflammation on the hepatic regulation of apoA-I, PON-1 and SAA in vivo, C57BL/6 mice received an intraperitoneal (ip.) injection of lipopolysaccharide (LPS) (0 to 100 µg) and liver tissue was collected 24 hour later.

To study the role of PPAR in the regulation of apoA-I, PON-1, and SAA, wild-type C57BL/6 (WT) and PPAR +/+ (129S3/SvImJ) male mice were obtained from The Jackson Laboratories, Bar Harbor, Me. PPAR -/- mice were a generous gift from Dr F. Gonzalez (National Institutes of Health, Bethesda, Md).

To study the role of NF- B in the regulation of these genes in vivo, double and single transgenic mice were obtained from crosses of homozygous mice expressing GLVP. GLVP is a chimeric transcriptional activator that consists of a mutated human progesterone receptor ligand binding domain fused to the GAL4 DNA binding and the HSV VP16 transcriptional activation domains, under the control of the liver-specific transthyretin promoter. Single transgenic mice (G mice) express the GLVP (G) transgene and double transgenic mice (GN mice) express both the GLVP (G) transgene and N-I B transgene (N) under the control of the GAL4 promoter, as described previously. 33 In these experiments, 5 mg/kg mifepristone (dissolved in sesame seed oil) was injected ip 3 hours before injection of TNF- (25 µg/kg mouse) and liver tissue was harvested 24 hours later. Mifepristone binds to GLVP, which can induce the expression of the nonphosphorylatable N-terminal deletion mutant of I B ( N-I B ), which inhibits NF- B activation in double transgenic mice. For controls, single GLVP transgenic mice (G) received mifepristone and TNF- or saline, and double transgenic mice (GN) received mifepristone and saline, or sesame seed oil.

Northern Blot and Western Blot Analysis

RNA (20 µg) was separated on an 1% agarose formaldehyde gel and transferred to Zeta Probe GT membranes (BioRad). Blots were hybridized with [ 32 P]-labeled cDNA probes of SAA1, apoA-I, and PON-1 (a generous gift from Dr Clem Furlong, University of Washington 34 ), and analyzed using a Cylcone-phosphorimager with Optiquant software (Packard). A GAPDH probe was used to normalize expression.

Hepa 1-6 cells were treated for 72 hours with the cytokine mix and/or other reagent(s). After incubation, cultured media were collected, concentrated, and then separated by 10 to 20% gradient SDS-PAGE and analyzed by Western blot using anti-SAA antibodies (a generous gift from Dr Godfrey Getz, University of Chicago, Ill), or anti-apoA-I polyclonal antibodies (Biodesign).

Multiplex Real-Time Quantitative Reverse-Transcription Polymerase Chain Reaction

Real-time reverse-transcription polymerase chain reaction (RT-PCR) was performed with the use of the TaqMan Master kit (Applied Biosystems) in the Stratagene MX3000P system. SAA1 primers and a FAM probe were obtained from Applied Biosystems (Assay-on-Demand). Primers and TaqMan probes specific for apoA-I, PON-1, and GAPDH are as follows: for apoA-1, forward primer, 5'ACCCCAGACTGTCGGAGAGC3'; reverse primer, 5'CCATTTTCCAAAGGTTTATT3'; probe, Cy5-5'CTGAACCTGAATCTCCTGG3'-BHQ2; for PON-1, forward primer, 5'GCATCTGAAAACCATCACACA3'; reverse primer, 5'AAGCTCTCAGGTCCAATAGCA3'; probe, ROX-5'CTTCTGCCTAGCATCAACGA3'-BHQ2; for GAPDH, forward primer, 5'AGCCTCGTCCCGTAGACAAA3'; reverse primer, 5'ACCAGGCGCCCAATACG3'; probe, HEX-5'AAATCCGTTCACACCGACCTTCACCA3'-BHQ1. Each sample was analyzed in triplicate and normalized in multiplex reactions using GAPDH as control.

Electromobility Shift Assay

Nuclear extracts from PPAR +/+ and PPAR -/- mouse hepatocytes were isolated and analyzed by electromobility shift assay (EMSA). 35 The NF- B double stranded consensus oligonucleotide was end-labeled with [ 32 P]-ATP. Nuclear extracts (10 µg) were incubated with the labeled NF- B oligonucleotide and 0.1 µg/µL poly (dI/dC). The DNA-protein complexes were separated electrophoretically and autoradiographed at -80°C. To assess specificity, competition assays were performed with an excess of unlabeled NF- B oligonucleotide (cold probe) or mutant NF- B oligonucleotide (G C substitution, 5'-AGTTGAGG C GACTTTCCCAGGC-3') added to the binding reaction.

Results

Cytokine Stimulation of Hepatocytes Increases SAA Expression While Simultaneously Decreasing ApoA-I and PON-1 Expression

Cytokines such as IL-6, IL-1ß, and TNF- increase the hepatic expression of SAA. 7 Because inflammation reduces HDL concentrations, 13-15 we examined the effect of cytokines on the simultaneous secretion of SAA and apoA-I by cultured hepatocytes. Because in vivo conditions involve exposure to multiple cytokines, the murine hepatoma cell line Hepa 1-6 was stimulated with a mixture of cytokines known to be key signaling molecules in atherosclerosis, namely IL-6, IL-1ß, and TNF-. Cytokine stimulation significantly increased SAA protein, detected by immunoblotting, in the media of Hepa 1-6 cells, whereas the amount of apoA-I simultaneously decreased ( Figure 1 A). A suitable antibody was not available for PON-1. Consistent with the changes in protein levels, cytokine stimulation increased SAA mRNA expression in Hepa 1-6 cells in a time-dependent fashion. Conversely, apoA-I and PON-1 expression were both decreased in Northern blot analysis of Hepa 1-6 cells ( Figure 1 B). Similar effects were observed in both AML12 and NMH hepatocytes, with cytokine stimulation increasing SAA expression and decreasing expression of apoA-I and PON-1 (supplemental Figure I, available online at http://atvb.ahajournals.org).

Figure 1. Cytokines upregulate SAA while simultaneously downregulating apoA-I and PON-1 in hepatocytes. Hepa 1-6 cells were treated with a mixture of cytokines (10 ng/mL TNF-, 10 ng/mL IL-1, 10 ng/mL IL-6). A, After 72-hour incubation, culture media from each treatment were analyzed by immunoblot using specific antibodies against SAA and apoA-I. Total RNA was extracted from Hepa 1-6 (B) or AML12 cells (C) with or without exposure to a mixture of cytokines for the indicated time periods, and subjected to Northern blot analysis using SAA, apoA-I, and PON-1 cDNAs as probes (B), or analyzed for the quantification by multiplex real-time RT-PCR using SAA1 ( ), apoA-I ( ), and PON-1 () specific primers and probes, normalized by GAPDH (C). Values shown are means±SD. D, LPS was injected intraperitoneally into C57BL/6 mice at the indicated dose. After 24 hours, total RNA from mouse liver was isolated for Northern blot analysis. Each lane represents a single mouse. A cDNA encoding GAPDH was used as a control for loading equivalence of RNA.

To quantify the changes in SAA, apoA-I, and PON-1 expression induced by cytokines, the mRNA expression level of each gene was analyzed by real-time RT-PCR after stimulation of AML12 cells with cytokines. Consistent with the Northern blot analysis, SAA versus apoA-I and PON-1 expression changed inversely in a time-dependent fashion ( Figure 1 C), although AML12 cells respond more rapidly to cytokines than Hepa 1-6 (compare Figure 1B and 1 C). Each cytokine regulated SAA, apoA-I, and PON-1 expression independently. However, a combination of all three cytokines showed more potent effects and was used for all further experiments 7 (supplemental Figure II).

To determine whether inflammation resulted in similar changes in vivo, LPS (0 to 100 µg/mouse) was injected ip in C57BL/6 mice for hepatic mRNA analysis by Northern blot 24 hours later. Consistent with the in vitro data, SAA expression increased, whereas apoA-I and PON-1 expression decreased dose-dependently ( Figure 1 D). To exclude a direct effect of LPS on the regulation of these genes in hepatocytes, LPS was added directly to AML12 cells for 24 hours. Neither SAA or apoA-I expression levels were changed by LPS treatment, while the mixture of cytokines regulated these genes (supplemental Figure III). These observations imply that cytokines produced in response to LPS stimulation of nonprarenchymal liver cells, in particular, Kupffer cells or extrahepatic macrophages, were stimulated by LPS in vivo, which in turn activated the hepatocytes. These results are consistent with previous observations. 36,37

To investigate whether the mitogen-activated protein (MAP) kinase signal transduction pathway was involved in these cytokine-mediated effects on the SAA, apoA-I, and PON-1 genes, the MEK inhibitors, PD98059 (50 µmol/L) and U0126 (10 µmol/L), which inhibit extracellular signal regulated kinase (ERK) pathway, were added to AML12 cells in the presence and absence of cytokines. Neither inhibitor attenuated the effect of the cytokines on these genes (data not shown).

NF- B Inhibitors Block the Cytokine Effects on SAA, ApoA-I, and PON-1 mRNA Levels

NF- B binding sites are present on the promoters of the acute phase form of SAA produced by liver, SAA1 and 2, 7 but not on the promoters of either apoA-I or PON-1. 25,27 To evaluate the involvement of the NF- B pathway in the cytokine-induced changes on SAA, apoA-I, and PON-1 expression, the NF- B inhibitors PDTC (a universal NF- B inhibitor) and SN50 (a peptide inhibitor of NF- B translocation) were added to Hepa 1-6 cells concurrently with the cytokine mix. PDTC partially inhibited cytokine-induced SAA secretion and partially restored apoA-I secretion ( Figure 2 A). Moreover, SN50 partially inhibited the increase of SAA mRNA expression and partially reversed the suppression of apoA-I and PON-1 mRNA expression evoked by the cytokines ( Figure 2 B).

Figure 2. NF- B inhibitors antagonize the alteration of SAA, apoA-I, and PON-1 induced by cytokines. A, Culture media from Hepa 1-6 cells incubated for 72 hours with a mixture of cytokines and/or the NF- B inhibitor PDTC were analyzed by immunoblot using specific antibodies against SAA and apoA-I. B, Total RNA was extracted from Hepa 1-6 cells treated with a mixture of cytokines and the NF- B inhibitor SN50 for 24 hours for Northern blot analysis.

The I B Super-Repressor Inhibits the Reciprocal Effect of Cytokines on SAA, and ApoA-I and PON-1

I B is a well-established endogenous repressor of NF- B. To further study the role of NF- B in the effects of cytokines on SAA, apoA-I, and PON-1, a super-repressor expression vector for I B was transiently transfected into Hepa 1-6 cells. CMV- N-I B express an I B super-repressor under the control of the cytomegalovirus (CMV) promoter, where the N-terminal deletion mutant of I B prevents the phosphorylation that is necessary for it to dissociate from NF- B. 30 Overexpression of the I B super-repressor by transient transfection into Hepa 1-6 cells partially inhibited the expression of SAA and ameliorated the decrease of apoA-I and PON-1 induced by cytokines ( Figure 3 ).

Figure 3. An I B super-repressor antagonizes the alteration of SAA, apoA-I, and PON-1 induced by cytokines. Hepa 1-6 cells were transiently transfected with the I B super-repressor expression vector or a mock control vector, as indicated. After transfection, cells were treated with a mixture of cytokines for 24 hours. Total RNA was extracted for Northern blot analysis.

To confirm the role of NF- B in vivo, we used GLVP/ N-I B transgenic mice, where the I B super-repressor is expressed in 50% of hepatocytes after the injection of mifepristone, 33 which selectively turns on the I B super-repressor. Thus, in double transgenic GLVP/ N-I B mice (GN), mifepristone induces the I B super-repressor, resulting in partial inhibition of NF- B transactivation. In contrast, in single transgenic GLVP transgenic mice (G), NF- B transactivation is normal because of the lack of expression of I B super-repressor ( N-I B ), regardless of mifepristone administration. Thus, single transgenic mice (G), with normal NF- B transactivation, were used as controls versus the double transgenic mice (GN), in which NF- B transactivation was inhibited. Three hours after injection of mifepristone (5 mg/kg) to induce hepatic expression of the I B super-repressor, TNF- (25 µg/kg) was injected for 24 hours before harvesting liver for Northern blot analysis ( Figure 4 ). Controls included (1) single transgenic mice (G) treated with mifepristone (lane 2), (2) double transgenic mice (GN) treated without mifepristone treatment (lanes 3 and 4), and (3) double transgenic mice (GN) treated with mifepristone but without TNF- (lanes 5 and 6). Expression of SAA, apoA-I, and PON-1 did not differ between any of these control mice (compare lanes 1-6). In contrast, SAA expression was markedly reduced by the super-repressor I B in the double transgenic mice (GN) injected with mifepristone (and thus expressing N-I B ) and TNF- (lanes 10 to 12), compared with single transgenic mice (that lack N-I B ) injected with mifepristone and TNF- (lanes 7 to 9). Moreover, apoA-I and PON-1 expression levels in the double transgenic mice (GN) were restored to that in the control mice (compare lanes 7 to 9 with 10 to 12), although SAA did not return fully to the basal state (compare lanes 4 to 6 with 10 to 12).

Figure 4. An I B super-repressor antagonizes the alterations of SAA, apoA-I, and PON-1 induced by injection of TNF- in GLVP/ N-I B transgenic mice. Single (G) and double (GN) transgenic mice were injected with mifepristone (5 mg/kg) or sesame seed oil for 3 hours before injection with TNF- (25 µg/kg) or saline as indicated. Liver tissue was harvested before TNF- injection (control or 0 hour) or 24 hours after TNF- injection. Total RNA was prepared and Northern blot analysis was performed. Each lane represents a single mouse.

PPAR Agonists Reverse the Effect of Cytokines on SAA, ApoA-I, and PON-1 Expression

The ligand-activated transcription factor PPAR limits inflammation via effects on NF- B activation. 38,39 Moreover, PPAR can induce apoA-I expression. Therefore, we examined the effects of the PPAR ligands WY-14643 and fenofibrate (CalBiochem) on cytokine-mediated expression of apoA-I and PON-1, both of which are known to have PPRE in their promoters, 25,27 and SAA, which does not. WY-14643 blocked the cytokine-mediated secretion of SAA and restored apoA-I secretion ( Figure 5 A). Both WY-14643 and fenofibrate reversed these cytokine effects, inhibiting the increase in SAA mRNA expression and ameliorating the suppression of apoA-I and PON-1 mRNA expression ( Figure 5 B). Thus, PPAR activation antagonizes cytokine-induced NF- B activation in hepatocytes, increasing apoA-I and PON-1 and decreasing SAA expression. However, in the absence of cytokines, fenofibrate did not increase apoA-I mRNA expression in either cell lines, whereas WY-14643 led to a slight increase in apoA-I mRNA.

Figure 5. PPAR agonists block the alterations of SAA, apoA-I, and PON-1 induced by cytokines. Hepa 1-6 and AML12 cells were treated with the PPAR ligands, WY-14643 (20 to 100 µmol/L), fenofibrate (100 µm), and/or a mixture of cytokines. A, After incubation for 72 hours, culture media were analyzed by immunoblot. The intensity of the apoA-I bands (lower panel) was measured and plotted; the band intensity from untreated cells was set at 100%. Values shown are means±SD. B, Total RNA after 24-hour exposure was analyzed by Northern blot.

Because both WY-14643 and fenofibrate could potentially act independently of PPAR, the effects of these PPAR agonists was tested in hepatocytes isolated from PPAR -deficient mice. Primary cultures of hepatocytes from PPAR +/+ and PPAR -/- mice were exposed to cytokines either with or without WY-14643 (100 µmol/L) or fenofibrate (100 µmol/L). Interestingly, hepatocytes from PPAR -/- mice expressed high levels of SAA and low level of apoA-I and PON-1 at baseline as compared with wild-type animals ( Figure 6 ). Neither WY-14643 nor fenofibrate had any effect on SAA, apoA-I, or PON-1 in PPAR -/- mice, either in the presence or absence of cytokines ( Figure 6 ). In primary hepatocytes from the wild-type control mice, the effect on SAA, apoA-I, and PON-1 was the same as seen in the other cell types ( Figure 6 A). Also, neither WY-14643 nor fenofibrate alone upregulated apoA-I expression. To investigate the NF- B transactivation in both PPAR +/+ and PPAR -/- mouse livers, EMSA was performed ( Figure 6 B). Compared with PPAR +/+ mice, the level of NF- B transactivation was high in PPAR -/- mice in the basal state (compare lane 5 with lane 1). Treatment with WY-14643 blocked NF- B transactivation in PPAR +/+ mice, whereas treatment with WY-14643 in PPAR -/- mice did not inhibit NF- B transactivation (compare lanes 3 and 4 with lanes 7 and 8). These gel shift bands appear specific because formation of the complex was blocked with either an unlabeled oligonucleotide, but not a mutant NF- B oligonucleotide. These results suggest that PPAR agonists limit cytokine-induced changes in SAA, apoA-I, and PON-1 in a PPAR -dependent manner, perhaps by modulating NF- B activation.

Figure 6. A, Cytokine regulation of SAA, apoA-I, and PON-1 gene expression by hepatocytes from PPAR +/+ and PPAR -/- mice. Primary hepatocytes cultured from PPAR +/+ and PPAR -/- mice were exposed to a cytokine mixture in the presence or absence of the PPAR agonists, fenofibrate or WY-14643. After 24 hours, total RNA was extracted from cells and analyzed by Northern blot. B, WY-14643 inhibits cytokine-induced activation of NF- B in hepatocytes via a PPAR -dependent mechanism. PPAR +/+ and PPAR -/- mouse hepatocytes were treated with WY-14643 (100 µmol/L) and/or a mixture of cytokines (TNF-, IL-1, and IL-6, all at 10 ng/mL) for 24 hours. Nuclear extracts (10 µg) were analyzed by EMSA using 100 ng of radiolabeled NF- B oligonucleotide. Specificity was determined by the addition of unlabeled NF- B (cold probe) or mutant NF- B oligonucleotide before the addition of the labeled probe. WY-14643 inhibited cytokine-induced activation of NF- B in PPAR +/+ but not in PPAR -/- hepatocytes.

To determine whether a mixture of cytokines can regulate either NF- B and PPAR expression, the protein levels of these nuclear transcription factors in lysates of cells treated with or without cytokines were examined by Western blot. After 24-hour exposure to cytokines, the expression levels of both nuclear factor did not demonstrate much change (supplemental Figure IV).

Discussion

We have demonstrated that cytokine stimulation of hepatocytes coordinately and reciprocally regulates critical HDL apolipoproteins, increasing SAA and repressing both apoA-I and PON-1 expression. These changes were observed in 3 murine hepatocytes cell lines, in primary cultures of mouse hepatocytes, and in mouse liver in vivo after LPS injection. These cytokine-induced changes all were reversed by decreasing NF- B activation or by PPAR agonists. Moreover, in the absence of the PPAR gene, isolated primary murine hepatocytes displayed a pattern of expression of these 3 HDL apolipoproteins similar to that seen after inflammatory stimuli.

Migita et al showed that stimulation of SAA by IL-1ß in primary human hepatocytes was in large part related to activation of MAP kinase rather than NF- B. 40 Although these data differ from ours, their experiments were performed in human liver cells, in which no SAA response was observed to either TNF- or IL-6, which is different to what we observed in murine hepatocytes. Moreover, in our experiments using a mixture of cytokines (IL-1ß, TNF-, and IL-6), the IL-1ß effect represented only a small percentage of the response to the cytokine mixture used in AML-12 cells (supplemental Figure I). Although it is conceivable that MAP-kinase activation is playing a minor role in our system, which we are unable to detect by the use of MEK inhibitors, our findings using both chemical inhibitors and molecular approaches in vitro and our in vivo studies in mice indicate a critical role for NF- B in the SAA response to cytokines.

This reciprocal regulation of SAA versus apoA-I and PON-1 appears to involve an interaction between PPAR and NF- B. Although the apoA-I and PON-1 promoters do not have NF- B sites, NF- B inactivation has been shown to reverse LPS-induced suppression of apoA-I expression in a human hepatoma cell line. 17 Moreover, a selective PPAR inhibitor blocked the increased apoA-I expression seen after overexpressing an I B super-repressor, suggesting that NF- B repression induces apoA-I expression through a PPAR -mediated mechanism. 17 Our data indicate a similar effect for PON-1 and reciprocal finding for SAA.

Our findings extend insight into the transcriptional regulation of HDL apolipoprotein through PPARs and NF- B. Suppression of NF- B activation with either chemical inhibitors or molecular approaches partially reversed the cytokine effects on the expression of SAA, which has NF- B sites in its promoter, and on the expression of apoA-I and PON-1, which do not. We also demonstrate that the PPAR effects seen after LPS stimulation also apply to inflammatory cytokines, endogenous stimuli thought to promote atherosclerosis in vivo. We show that these effects extend to reciprocal and coordinated effects on PON-1 and SAA, and that all these effects require the genetic presence of PPAR. The tight, consistent, and coordinated reciprocal downregulation of apoA-I and PON-1 versus SAA upregulation during inflammation argues for a common mechanism involving cross-talk between PPAR and NF- B. Although SAA does not have a PPRE in its promoter, its expression was increased in the basal state in hepatocytes from PPAR -deficient mice and the cytokine-induced expression of SAA was reduced by exposure of Hepa 1-6, AML12, and NMH hepatocytes to 2 different PPAR agonists. We plan to examine the effect of these transcription factors on the promoter activities of SAA, apoA-I, and PON-1 in future studies.

There are several possible mechanisms by which inflammation could affect both NF- B and PPAR transcriptional activity. One mechanism would be regulation of the expression level of PPAR by cytokines. Others have shown that inflammation can repress PPAR expression, 41 although we did not find a difference in the amount of PPAR protein. Another mechanism would be through direct interaction between PPAR and NF- B. This will be the subject of future studies.

Our data indicate an important role for PPAR in both limiting inflammation under basal conditions as well as countering proinflammatory stimuli. The elevated SAA and suppressed apoA-I and PON-1 levels in the unstimulated state in the absence of PPAR indicate that the cells were in an inflammatory state, implicating PPAR as a "brake" that limits inflammation even in the basal state. Further evidence for a role for PPAR in suppressing basal levels of inflammation comes from the observation that endothelial cell expression of adhesion molecules, an indicator of endothelial inflammation, is markedly increased in PPAR -deficient mice. 42 PPAR -deficient mice also have prolonged responses to inflammatory stimuli. 43,44 The exact mechanism for these effects remains unclear, because SAA lacks a PPRE in its promoter region. Regardless, these observations suggest cross-talk between NF- B, activated by cytokines, (inflammation) and PPAR, activated by ligands (fibrates), affecting the expression of genes other than those directly activated by either NF- B or PPAR. Thus, NF- B activated by cytokines can influence the expression of apoA-I and PON-1, and ligand activation of PPAR can regulate the expression of SAA.

The observation that cytokines suppressed apoA-I expression by PPAR agonists in murine Hepa 1-6 cells was somewhat unexpected because the PPRE in rodent apoA-I has been reported to be mutated and nonresponsive to PPAR agonists. 26 Nonetheless, experiments in 3 different murine hepatocyte cell lines using 2 different PPAR agonists, WY-14643 and fenofibrate, yielded similar results. Moreover, the changes in apoA-I expression was absent in hepatocytes from PPAR -deficient mice. Consistent with findings of Staels, 26,45 we also failed to show a consistent effect of fibrates on apoA-I expression in the absence of cytokine stimulation. It is possible that the effect of these agents on apoA-I expression is occurring indirectly through a PPAR -independent mechanism. However, the findings in hepatocytes from the PPAR -deficient mice strongly support some involvement of this nuclear receptor in the regulation of both SAA and apoA-I expression, even if indirect. Human apoA-I expression is known to be PPAR -regulated. 25,26 Our results in the PPAR -deficient animals suggest that PPAR does affect the expression of mouse apoA-I, and that PPAR activation can mitigate NF- B activation by inflammatory stimuli.

In vitro studies suggest that SAA can displace apoA-I from HDL particles. 4,23 Our findings suggest an alternate explanation for the reduction in HDL levels seen during inflammation. Reduced expression of apoA-I by hepatocytes, the primary site of synthesis of HDL, with simultaneous increased SAA expression, would result in the secretion of HDL particles of altered composition. The simultaneous reduction in PON-1 expression would also contribute to alterations in HDL structure during inflammation, with likely functional consequences.

Inflammatory HDL has been shown to have a reduced capacity to inhibit LDL oxidation and the adhesion of monocytes to endothelial cells than control HDL, 20,22 which has in part been attributed to the well-documented reduction in the PON-1 content of HDL that occurs during inflammation. 21,22 Thus, PON-1 has been regarded as being atheroprotective, a hypothesis reinforced by the increased atherosclerosis that occurs in the PON-1-deficient mouse 46 and the decreased atherosclerosis seen in the PON-1 transgenic mouse. 47 The major atheroprotective apolipoprotein in HDL has been thought to be apoA-I, which plays a critical role in reverse cholesterol transport. 48 ApoA-I transgenic mice are protected against atherosclerosis, 49,50 whereas atherosclerotic lesions are increased in apoA-I-deficient mice. 51 Thus, inflammation-mediated reduction in the apoA-I content of HDL would be predicted to reduce the atheroprotective effect of HDL.

The role of SAA in atherogenesis remains unclear. Despite epidemiological associations between circulating SAA levels and cardiovascular risk in humans, 1,2 there is no definitive evidence implicating SAA as having a direct role in atherogenesis or cardiovascular disease. However, we recently demonstrated that diet-induced increases in SAA levels were associated with an increase in atherosclerosis in LDL receptor-deficient mice, independent of changes in plasma lipids and lipoproteins. Additionally, we showed that HDL binding to vascular proteoglycans in vitro was proportional to the SAA content of the lipoprotein and that SAA, apoA-I, and proteoglycans colocalized in lesions in these mice. 6 These findings are consistent with, but do not prove, the notion that SAA is directly involved in atherogenesis by tethering HDL to vascular proteoglycans.

In summary, the present findings suggest a novel mechanism through which atheroprotective HDL apolipoproteins (apoA-I and PON-1) and the inflammatory and potentially atherogenic HDL apolipoprotein, SAA, are reciprocally regulated by inflammation. We propose a model whereby these reciprocal changes during cytokine-mediated inflammation are regulated by an interaction between PPAR and NF- B, inducing counter-regulatory transcriptional responses through changes in expression of their target genes. Moreover, our findings provide further evidence for PPAR expression exerting a chronic "braking" effect on inflammation, which can be reversed by inflammatory cytokines or the absence of PPAR itself.

Acknowledgments

Sources of Funding

This work was supported by grants from the National Institutes of Health (HL30086, HL18645, HL079117, HL071745, HL048743, CA23226, and CA074131) and a grant from the Donald W. Reynolds Foundation.

Disclosure(s)

None.

【参考文献】
  Ridker PM, Hennekens CH, Buring JE, Rifai N. C-reactive protein and other markers of inflammation in the prediction of cardiovascular disease in women. N Engl J Med. 2000; 342: 836-843.

Libby P. Inflammation in atherosclerosis. Nature. 2002; 420: 868-874.

Eriksen N, Benditt EP. Isolation and characterization of the amyloid-related apoprotein (SAA) from human high density lipoprotein. Proc Natl Acad Sci U S A. 1980; 77: 6860-6864.

Coetzee GA, Strachan AF, van der Westhuyzen DR, Hoppe HC, Jeenah MS, de Beer FC. Serum amyloid A-containing human high density lipoprotein 3. Density, size, and apolipoprotein composition. J Biol Chem. 1986; 261: 9644-9651.

Ancsin JB, Kisilevsky R. Serum amyloid A peptide interactions with glycosaminoglycans. Evaluation by affinity chromatography. Methods Mol Biol. 2001; 171: 449-456.

Lewis KE, Kirk EA, McDonald TO, Wang S, Wight TN, O?Brien KD, Chait A. Increase in serum amyloid a evoked by dietary cholesterol is associated with increased atherosclerosis in mice. Circulation. 2004; 110: 540-545.

Uhlar CM, Whitehead AS. Serum amyloid A, the major vertebrate acute-phase reactant. Eur J Biochem. 1999; 265: 501-523.

Fruchart J-C, Ailhaud G, Denefle P. Apo A-containing lipoprotein particles: Interactions with HDL binding sites and reverse cholesterol transport. In: Miller NE, Tall AR, eds. High Density Lipoproteins and Atherosclerosis III. Amsterdam: Elsevier Science; 1992.

Segrest JP, Li L, Anantharamaiah GM, Harvey SC, Liadaki KN, Zannis V. Structure and function of apolipoprotein A-I and high-density lipoprotein. Curr Opin Lipidol. 2000; 11: 105-115.

Denis M, Haidar B, Marcil M, Bouvier M, Krimbou L, Genest J Jr. Molecular and cellular physiology of apolipoprotein A-I lipidation by the ATP-binding cassette transporter A1 (ABCA1). J Biol Chem. 2004; 279: 7384-7394.

Oram J, Lawn R. ABCA1: The gatekeeper for eliminating excess tissue cholesterol. J Lipid Res. 2001; 42: 1173-1179.

Ly H, Francone OL, Fielding CJ, Shigenaga JK, Moser AH, Grunfeld C, Feingold KR. Endotoxin and TNF lead to reduced plasma LCAT activity and decreased hepatic LCAT mRNA levels in Syrian hamsters. J Lipid Res. 1995; 36: 1254-1263.

Tietge UJ, Maugeais C, Cain W, Rader DJ. Acute inflammation increases selective uptake of HDL cholesteryl esters into adrenals of mice overexpressing human sPLA2. Am J Physiol Endocrinol Metab. 2003; 285: E403-E411.

Cabana VG, Lukens JR, Rice KS, Hawkins TJ, Getz GS. HDL content and composition in acute phase response in three species: triglyceride enrichment of HDL a factor in its decrease. J Lipid Res. 1996; 37: 2662-2674.

Pruzanski W, Stefanski E, de Beer FC, de Beer MC, Ravandi A, Kuksis A. Comparative analysis of lipid composition of normal and acute-phase high density lipoproteins. J Lipid Res. 2000; 41: 1035-1047.

Fredrikson GN, Hedblad B, Nilsson JA, Alm R, Berglund G, Nilsson J. Association between diet, lifestyle, metabolic cardiovascular risk factors, and plasma C-reactive protein levels. Metabolism. 2004; 53: 1436-1442.

Morishima A, Ohkubo N, Maeda N, Miki T, Mitsuda N. NF- B regulates plasma apolipoprotein A-I and high density lipoprotein cholesterol through inhibition of peroxisome proliferator-activated receptor alpha. J Biol Chem. 2003; 278: 38188-38193.

Haas MJ, Horani M, Mreyoud A, Plummer B, Wong NC, Mooradian AD. Suppression of apolipoprotein AI gene expression in HepG2 cells by TNF and IL-1ß. Biochim Biophys Acta. 2003; 1623: 120-128.

Bergmeier C, Siekmeier R, Gross W. Distribution Spectrum of Paraoxonase Activity in HDL Fractions. Clin Chem. 2004; 50: 2309-2315.

Van Lenten BJ, Hama SY, de Beer FC, Stafforini DM, McIntyre TM, Prescott SM, La Du BN, Fogelman AM, Navab M. Anti-inflammatory HDL becomes pro-inflammatory during the acute phase response. Loss of protective effect of HDL against LDL oxidation in aortic wall cell cocultures. J Clin Invest. 1995; 96: 2758-2767.

Feingold KR, Memon RA, Moser AH, Grunfeld C. Paraoxonase activity in the serum and hepatic mRNA levels decrease during the acute phase response. Atherosclerosis. 1998; 139: 307-315.

Van Lenten BJ, Wagner AC, Nayak DP, Hama S, Navab M, Fogelman AM. High-density lipoprotein loses its anti-inflammatory properties during acute influenza a infection. Circulation. 2001; 103: 2283-2288.

Husebekk A, Skogen B, Husby G. Characterization of amyloid proteins AA and SAA as apolipoproteins of high density lipoprotein (HDL). Displacement of SAA from the HDL-SAA complex by apo AI and apo AII. Scand J Immunol. 1987; 25: 375-381.

Cabana VG, Reardon CA, Feng N, Neath S, Lukens J, Getz GS. Serum paraoxonase: effect of the apolipoprotein composition of HDL and the acute phase response. J Lipid Res. 2003; 44: 780-792.

Staels B, Auwerx J. Regulation of apo A-I gene expression by fibrates. Atherosclerosis. 1998; 137 Suppl: S19-23.

Berthou L, Duverger N, Emmanuel F, Langouet S, Auwerx J, Guillouzo A, Fruchart JC, Rubin E, Denefle P, Staels B, Branellec D. Opposite regulation of human versus mouse apolipoprotein A-I by fibrates in human apolipoprotein A-I transgenic mice. J Clin Invest. 1996; 97: 2408-2416.

Gouedard C, Koum-Besson N, Barouki R, Morel Y. Opposite regulation of the human paraoxonase-1 gene PON-1 by fenofibrate and statins. Mol Pharmacol. 2003; 63: 945-956.

Gouedard C, Barouki R, Morel Y. Induction of the paraoxonase-1 gene expression by resveratrol. Arterioscler Thromb Vasc Biol. 2004; 24: 2378-2383.

Wu JC, Merlino G, Cveklova K, Mosinger B Jr, Fausto N. Autonomous growth in serum-free medium and production of hepatocellular carcinomas by differentiated hepatocyte lines that overexpress transforming growth factor alpha 1. Cancer Res. 1994; 54: 5964-5973.

Scatena M, Almeida M, Chaisson ML, Fausto N, Nicosia RF, Giachelli CM. NF- B mediates vß3 integrin-induced endothelial cell survival. J Cell Biol. 1998; 141: 1083-1093.

Jacobsen LB, Calvin SA, Colvin KE, Wright M. FuGENE 6 Transfection Reagent: the gentle power. Methods. 2004; 33: 104-112.

Klaunig JE, Goldblatt PJ, Hinton DE, Lipsky MM, Trump BF. Mouse liver cell culture. II. Primary culture. In Vitro. 1981; 17: 926-934.

Chaisson ML, Brooling JT, Ladiges W, Tsai S, Fausto N. Hepatocyte-specific inhibition of NF- B leads to apoptosis after TNF treatment, but not after partial hepatectomy. J Clin Invest. 2002; 110: 193-202.

Li WF, Matthews C, Disteche CM, Costa LG, Furlong CE. Paraoxonase (PON1) gene in mice: sequencing, chromosomal localization and developmental expression. Pharmacogenetics. 1997; 7: 137-144.

Marx N, Sukhova GK, Collins T, Libby P, Plutzky J. PPAR activators inhibit cytokine-induced vascular cell adhesion molecule-1 expression in human endothelial cells. Circulation. 1999; 99: 3125-3131.

Scott MJ, Liu S, Su GL, Vodovotz Y, Billiar TR. Hepatocytes enhance effects of lipopolysaccharide on liver nonparenchymal cells through close cell interactions. Shock. 2005; 23: 453-458.

Matsumura T, Degawa T, Takii T, Hayashi H, Okamoto T, Inoue J, Onozaki K. TRAF6-NF- B pathway is essential for interleukin-1-induced TLR2 expression and its functional response to TLR2 ligand in murine hepatocytes Immunology. 2003; 109: 127-136.

Duval C, Chinetti G, Trottein F, Fruchart JC, Staels B. The role of PPARs in atherosclerosis. Trends Mol Med. 2002; 8: 422-430.

Ziouzenkova O, Asatryan L, Sahady D, Orasanu G, Perrey S, Cutak B, Hassell T, Akiyama TE, Berger JP, Sevanian A, Plutzky J. Dual roles for lipolysis and oxidation in peroxisome proliferation-activator receptor responses to electronegative low density lipoprotein. J Biol Chem. 2003; 278: 39874-39881.

Migita K, Miyashita T, Maeda Y, Nakamura M, Yatsuhashi H, Ishibashi H, Eguchi K. An active metabolite of leflunomide, A77 1726, inhibits the production of serum amyloid A protein in human hepatocytes. Rheumatology (Oxford). 2005; 44: 443-448.

Beigneux AP, Moser AH, Shigenaga JK, Grunfeld C, Feingold KR. The acute phase response is associated with retinoid X receptor repression in rodent liver. J Biol Chem. 2000; 275: 16390-16399.

Ziouzenkova O, Perrey S, Asatryan L, Hwang J, MacNaul KL, Moller DE, Rader DJ, Sevanian A, Zechner R, Hoefler G, Plutzky J. Lipolysis of triglyceride-rich lipoproteins generates PPAR ligands: evidence for an antiinflammatory role for lipoprotein lipase. Proc Natl Acad Sci U S A. 2003; 100: 2730-2735.

Devchand PR, Keller H, Peters JM, Vazquez M, Gonzalez FJ, Wahli W. The PPAR -leukotriene B4 pathway to inflammation control. Nature. 1996; 384: 39-43.

Blanquart C, Barbier O, Fruchart JC, Staels B, Glineur C. Peroxisome proliferator-activated receptors: regulation of transcriptional activities and roles in inflammation. J Steroid Biochem Mol Biol. 2003; 85: 267-273.

Vu-Dac N, Chopin-Delannoy S, Gervois P, Bonnelye E, Martin G, Fruchart JC, Laudet V, Staels B. The nuclear receptors peroxisome proliferator-activated receptor alpha and Rev-erbalpha mediate the species-specific regulation of apolipoprotein A-I expression by fibrates. J Biol Chem. 1998; 273: 25713-25720.

Shih DM, Xia YR, Wang XP, Miller E, Castellani LW, Subbanagounder G, Cheroutre H, Faull KF, Berliner JA, Witztum JL, Lusis AJ. Combined serum paraoxonase knockout/apolipoprotein E knockout mice exhibit increased lipoprotein oxidation and atherosclerosis. J Biol Chem. 2000; 275: 17527-17535.

Tward A, Xia YR, Wang XP, Shi YS, Park C, Castellani LW, Lusis AJ, Shih DM. Decreased atherosclerotic lesion formation in human serum paraoxonase transgenic mice. Circulation. 2002; 106: 484-490.

Oram JF, Mendez AJ, Bierman EL. HDL receptor-mediated transport of cholesterol from cells. In: Miller NE, Tall AR, eds. High Density Lipoprotein and Atherosclerosis. Amsterdam: Elsevier Science; 1992.

De Geest B, Stengel D, Landeloos M, Lox M, Le Gat L, Collen D, Holvoet P, Ninio E. Effect of overexpression of human apo A-I in C57BL/6 and C57BL/6 apo E-deficient mice on 2 lipoprotein-associated enzymes, platelet-activating factor acetylhydrolase and paraoxonase. Comparison of adenovirus-mediated human apo A-I gene transfer and human apo A-I transgenesis. Arterioscler Thromb Vasc Biol. 2000; 20: e68-e75.

Zhang Y, Zanotti I, Reilly MP, Glick JM, Rothblat GH, Rader DJ. Overexpression of apolipoprotein A-I promotes reverse transport of cholesterol from macrophages to feces in vivo. Circulation. 2003; 108: 661-663.

Hughes SD, Verstuyft J, Rubin EM. HDL deficiency in genetically engineered mice requires elevated LDL to accelerate atherogenesis. Arterioscler Thromb Vasc Biol. 1997; 17: 1725-1729.

作者单位：Departments of Medicine (C.Y.H., T.C., A.C.) and Pathology (J.S.C., N.F., M.C.), University of Washington, Seattle; and the Cardiovascular Division (G.O., J.P.), Brigham and Women?s Hospital, Harvard Medical School, Boston, Mass.